Development of a Dam Reservoir Seeking Model Applying Semantic Segmentation to Sentinel-2 Satellite Data

Key Points

• The aim is to create a model for exploring dam reservoirs using semantic segmentation applied to satellite data.
• Applied semantic segmentation to Sentinel-2 satellite data.
• Utilized existing datasets and image processing to create training data.
• Evaluated model performance using metrics like F1 Score and IoU.
• Achieved F1 Score of ≧0.60 and IoU of ≧0.54, indicating good model accuracy.
• The model maintained high accuracy when applied to datasets from China and Southeast Asia.